Upload File

chapman conference on the nao, spain, 28 variability of...

  • Home
  • Documents
  • Chapman Conference on the NAO, Spain, 28 Variability of ...xtide.ldeo.columbia.edu/~visbeck/nao/poster/Braesicke.pdf · choice of limiters; Superbee or Van Leer (Cullen and Barnes,
1
Chapman Conference on the NAO, Spain, 28 November - 1 December Variability of Total Ozone due to the NAO as Represented in two Different Model Simulations P. Braesicke, P. Hadjinicolaou, A. Jrrar and J. Pyle Centre for Atmospheric Science, Cambridge University, UK Model Descriptions and Experimental Design Abstract: In this study two model systems are assessed on how they represent total ozone variability due to the influence of the North Atlantic Oscillation (NAO). The main focus will be on the northern hemisphere winter and especially on January. A brief comparison with TOMS data will also be included. One model is the chemistry-transport model (CTM) SLIMCAT, which is driven by meteorological observations and therefore includes meteorological trends in the atmospheric circulation. The other is the Unified Model (UM) of the UK Met. Office, which is a comprehensive general circulation model forced by prescribed sea-surface temperatures (SSTs) only. In both models a simplified ozone chemistry (Cariolle and D´ equ´ e, 1986) is used with a param- eterization of polar ozone loss, but with no allowance for changing chlorine or aerosol levels on the chemistry. Suitable indices of the NAO will be used in the diagnosis of the two model systems and the correlation between these indices and the total ozone as de- rived by the models will be compared. Composite ozone maps will be con- structed for high and low index phases and will be also compared with TOMS observations. Unified Model: horizontal resolution 96 73 grid points (3.75 in longitude and 2.5 in latitude) vertical resolution 58 levels (L58); z=1.3 km, model top in 0.1 hPa (65 km) timestep 15 minutes (900 s) tracer transport Roe flux redistribution method with choice of limiters; Superbee or Van Leer (Cullen and Barnes, 1997) gravity wave drag based on Palmer et al. 1986 radiation 2-stream radiation code; 6 bands SW, 8 bands LW; every 3 hours (Edwards and Slingo, 1996) ozone Cariolle parameterization with an addi- tional ”cold tracer” (passive/interactive) ocean climatological SSTs (AMIP II, monthly mean data) SLIMCAT: horizontal resolution T21, 64 32 grid points (5.6 in longi- tude and 5.6 in latitude) vertical resolution 11 isentropic levels (L11); z=2.5 km, model top in 1030 K (8 hPa, 32 km), model bottom in 345 K (250 hPa, 10 km) timestep 1 hour (3600 s) tracer transport Second-order moments advection driven by ECMWF analysis (Prather, 1986; Gibson, 1997) radiation calculation of heating rates for the verti- cal transport; 2 dummy levels above and below model domain; uses climatologi- cal ozone (Shine, 1987; Shine and Rick- aby, 1989) ozone Cariolle parameterization with an addi- tional ”cold tracer” The NAO Indices in Comparison The plots in this box show the NAO indices for December, Jan- uary and February for the UM (left), SLIMCAT (middle) and from station data (right, after Hur- rell, 1995). The UM, only driven by prescribed AMIP II SSTs, dis- plays many differences compared to the other two time series. How- ever, there is good agreement for selected years (e.g. 1984). -4 -2 0 2 4 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 Index Time [Years] NAO Indices for the Unified Model December January February -4 -2 0 2 4 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 Index Time [Years] NAO Indices for SLIMCAT (ECMWF) December January February -4 -2 0 2 4 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 Index Time [Years] NAO Indices Derived from Observations December January February The gridpoint data used for SLIM- CAT (ECMWF, L31) shows good agreement to the station data, but does not capture the largest amplitudes. This is expected, because the nearest gridpoint is not only determined by the nearest meteorological station and is also influenced by the model system used for the assimilation. Correlations between NAO Indices and Total Ozone This box shows a comparison between the UM (left) and SLIMCAT (rght) correlations for the NAO index with total ozone as derived by the models for January and February. Focussing on the European/Atlantic section in January, both models show similar patterns: High positive correlations over Greenland and a band of nega- tive correlations over Europe, with positive cor- relations towards the southeast. The positive correlations over Greenland are stronger in the UM. In February the correlation over Greenland is stronger in SLIMCAT compared to the UM results but the patterns are again similar in both model systems. On the hemispheric scale the comparison is not that straightforward - away from the European/Atlantic sector many differ- ences can be seen, e.g. the smaller eastward extend of the band with positive correlations from Africa to Asia in January in SLIMCAT. Correlations between NAO Indices and Geopotential Heights Correlations between geopotential heights and the NAO index in January within the UM are discussed for three different pressure levels in this box: In 850 hPa a large area of negative correlations is found in polar latitudes, centred over Iceland and Greenland. The maximum positive correlations are as expected over Spain and North Africa - in accordance with the definition of the NAO index. In 200 hPa the pattern of the correlation resembles the pattern seen in the ozone (with opposite sign). This implies that the amount of total ozone over Europe is mainly controlled by the modulation of tropopause heights due to the NAO. In 30 hPa the correlation is not as high as in lower levels, and a dipole structure emerges. This may not only imply a strengthening/weakening of the vortex, but also a twisting of the Aleutian High/Vortex system. Similar results can be obtained for the SLIMCAT/ECMWF model system (except for 30 hPa, were a more zonal symmetric pattern appears). Composites of Ozone Anomalies By stratifying the monthly mean total ozone data according to the NAO indices and taking differ- ences between low and high in- dex cases, the composites on the right were derived. The general shape of the pattern is quite similar between models (UM, SLIMCAT) and observations (TOMS). SLIM- CAT produces the largest ampli- tudes and the strongest gradient around 55 N. The UM has a weaker gradient in this region, which is closer to observations. It seems to un- derestimate the negative anomaly over Europe. Also a small dis- placement to the north is obvious. The elongated positive anomaly over the Atlantic in the SLIM- CAT run is much larger than in the UM experiment. It seems likely that SLIMCAT overestimates this anomaly. Using only a simple stratification may still leave other signals (e.g. a linear trend) in the data (SLIMCAT and TOMS) - further work has to be done! Summary and Conclusions Both models show a similar response in total ozone due to their internal NAO. The response is mainly controlled in both model systems by the tropopause heights/topography of the 200 hPa surface. First comparisons with TOMS total ozone data show a good agree- ment between models (UM and SLIMCAT) and observations. Further work: How does the NAO variability affect the trends in total ozone?

Upload: others

Post on 28-Jan-2021

5 views

Category:

Documents


0 download

Report

TRANSCRIPT

  • �

    �

    �

    �

    Chapman Conference on the NAO, Spain, 28� �

    November - 1 ��December

    Variability of TotalOzonedueto theNAO asRepresentedin two DifferentModelSimulations

    P. Braesicke,P. Hadjinicolaou,A. JrrarandJ.PyleCentre for Atmospheric Science, Cambridge University, UK

    �

    �

    �

    �

    ModelDescriptionsandExperimentalDesignAbstract:In this study two model systemsare assessedon how they representtotalozonevariability dueto theinfluenceof theNorthAtlantic Oscillation(NAO).Themain focuswill beon thenorthernhemispherewinter andespeciallyonJanuary. A brief comparisonwith TOMS datawill alsobeincluded.One model is the chemistry-transportmodel (CTM) SLIMCAT, which isdrivenby meteorologicalobservationsandthereforeincludesmeteorologicaltrendsin the atmosphericcirculation. The otheris the Unified Model (UM)of theUK Met. Office, which is a comprehensive generalcirculationmodelforcedby prescribedsea-surfacetemperatures(SSTs)only. In bothmodelsasimplifiedozonechemistry(CariolleandDéqúe,1986)is usedwith a param-eterizationof polarozoneloss,but with no allowancefor changingchlorineor aerosollevelson thechemistry.Suitableindicesof the NAO will be usedin the diagnosisof the two modelsystemsandthecorrelationbetweentheseindicesandthe total ozoneasde-rived by the modelswill be compared.Compositeozonemapswill be con-structedfor highandlow index phasesandwill bealsocomparedwith TOMSobservations.

    Unified Model:horizontalresolution96 � 73 grid points (3.75� in longitude

    and2.5� in latitude)verticalresolution 58 levels (L58); z=1.3km, modeltop

    in 0.1hPa (65 km)timestep 15minutes(900s)tracertransport Roe flux redistribution method with

    choiceof limiters;Superbeeor VanLeer(CullenandBarnes,1997)

    gravity wavedrag basedonPalmeretal. 1986radiation 2-streamradiationcode;6 bandsSW, 8

    bandsLW; every 3 hours(EdwardsandSlingo,1996)

    ozone Cariolle parameterizationwith an addi-tional ”cold tracer”(passive/interactive)

    ocean climatologicalSSTs(AMIP II, monthlymeandata)

    SLIMCA T:horizontalresolutionT21, 64 � 32 grid points (5.6� in longi-

    tudeand5.6� in latitude)verticalresolution 11 isentropiclevels (L11); z=2.5 km,

    model top in 1030 K (8 hPa, 32 km),modelbottomin 345K (250hPa,10km)

    timestep 1 hour(3600s)tracertransport Second-ordermomentsadvectiondriven

    by ECMWF analysis (Prather, 1986;Gibson,1997)

    radiation calculationof heatingratesfor theverti-cal transport;2 dummylevelsabove andbelow modeldomain;usesclimatologi-cal ozone(Shine,1987;ShineandRick-aby, 1989)

    ozone Cariolle parameterizationwith an addi-tional ”cold tracer”

    �

    �

    �

    �

    TheNAO Indicesin ComparisonThe plots in this box show theNAO indices for December, Jan-uary and February for the UM(left), SLIMCAT (middle) andfrom stationdata(right, afterHur-rell, 1995). The UM, only drivenby prescribedAMIP II SSTs,dis-playsmany differencescomparedto theothertwo time series.How-ever, thereis goodagreementforselectedyears(e.g.1984).

    -4

    -2

    0

    2

    4

    1978 1980 1982 1984 1986 1988 1990 1992 1994 1996

    Inde

    x

    Time [Years]

    NAO Indices for the Unified Model

    DecemberJanuary

    February -4

    -2

    0

    2

    4

    1978 1980 1982 1984 1986 1988 1990 1992 1994 1996

    Inde

    x

    Time [Years]

    NAO Indices for SLIMCAT (ECMWF)

    DecemberJanuary

    February -4

    -2

    0

    2

    4

    1978 1980 1982 1984 1986 1988 1990 1992 1994 1996

    Inde

    x

    Time [Years]

    NAO Indices Derived from Observations

    DecemberJanuary

    February

    Thegridpointdatausedfor SLIM-CAT (ECMWF, L31) shows goodagreement to the station data,but does not capturethe largestamplitudes. This is expected,becausethe nearestgridpoint isnotonly determinedby thenearestmeteorologicalstationandis alsoinfluencedby the model systemusedfor theassimilation.

    �

    �

    �

    �

    CorrelationsbetweenNAO IndicesandTotalOzoneThis box shows a comparisonbetweentheUM(left) andSLIMCAT (rght) correlationsfor theNAO index with total ozoneasderived by themodelsfor JanuaryandFebruary. Focussingonthe European/Atlanticsectionin January, bothmodels show similar patterns: High positivecorrelationsoverGreenlandandabandof nega-tivecorrelationsoverEurope,with positivecor-relationstowards the southeast. The positivecorrelationsover Greenlandarestrongerin theUM.

    In Februarythe correlationover Greenlandisstronger in SLIMCAT comparedto the UMresultsbut thepatternsareagain similar in bothmodel systems. On the hemisphericscalethecomparisonis not that straightforward - awayfrom theEuropean/Atlanticsectormany differ-encescan be seen,e.g. the smallereastwardextend of the band with positive correlationsfrom Africa to Asia in Januaryin SLIMCAT.

    �

    �

    �

    �

    CorrelationsbetweenNAO IndicesandGeopotentialHeights

    CorrelationsbetweengeopotentialheightsandtheNAO index in Januarywithin theUM arediscussedfor threedifferentpressurelevelsin thisbox: In 850hPaalargeareaof negativecorrelationsis foundin polar latitudes,centredover IcelandandGreenland.Themaximumpositive correlationsareasexpectedover SpainandNorth Africa - in accordancewith thedefinitionof theNAO index. In 200hPathepatternof thecorrelationresemblesthepatternseenin theozone(with oppositesign). This implies that theamountof total ozoneover Europeis mainly controlledby themodulationof tropopauseheightsdueto theNAO. In 30 hPa thecorrelationis not ashigh asin lower levels,anda dipolestructureemerges.This maynot only imply a strengthening/weakeningof thevortex, but alsoa twisting oftheAleutianHigh/Vortex system.Similar resultscanbeobtainedfor theSLIMCAT/ECMWF modelsystem(exceptfor 30hPa,wereamorezonalsymmetricpatternappears).

    �

    �

    �

    �

    Compositesof OzoneAnomaliesBy stratifying the monthly meantotal ozonedataaccordingto theNAO indices and taking differ-encesbetweenlow and high in-dex cases,the compositeson theright were derived. The generalshapeof thepatternis quitesimilarbetweenmodels(UM, SLIMCAT)andobservations(TOMS). SLIM-CAT producesthe largest ampli-tudes and the strongestgradientaround55� N.

    The UM has a weaker gradientin this region, which is closer toobservations. It seemsto un-derestimatethe negative anomalyover Europe. Also a small dis-placementto thenorth is obvious.The elongated positive anomalyover the Atlantic in the SLIM-CAT runis muchlargerthanin theUM experiment. It seemslikelythat SLIMCAT overestimatesthisanomaly.

    Usingonly a simplestratificationmaystill leaveothersignals(e.g.a lineartrend)in thedata(SLIMCAT andTOMS) - furtherwork hasto bedone!

    �

    �

    �

    �

    SummaryandConclusions� Bothmodelsshow asimilar responsein totalozonedueto their

    internalNAO.

    � Theresponseis mainlycontrolledin bothmodelsystemsby thetropopauseheights/topography of the200hPasurface.

    � Firstcomparisonswith TOMStotalozonedatashow agoodagree-mentbetweenmodels(UM andSLIMCAT) andobservations.

    Furtherwork: How doestheNAO variability affect thetrendsin total ozone?


THOR CT3 Meeting – Torshavn 2009 – Fischer/Visbeck/Zantopp/Nunes

Nao julgue

Nao Se Iluda, Nao - Isabela Freitas (1)

Inverse Relationships Between NAO and Calanus Finmarchicus ...xtide.ldeo.columbia.edu/~visbeck/nao/poster/Conversi.pdf · NAO . C.f . abundance and NAO index are positively cross

Seasonality of the NAO Abstract Methodxtide.ldeo.columbia.edu/~visbeck/nao/poster/Portis.pdf · W H M M M M M M M M M M M M Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 0.2 0.0-0.2-0.4-0.6-0.8-1.0

Robot humanoïde NAO - roboteknologie.weebly.comroboteknologie.weebly.com/.../26804295/robot-humanoide-nao-erm-fr.pdfRobot humanoïde NAO Système d’étude de la robotique mobile

namibia nao

Nao Desista

Nao mataras

Atmospheric Properties II Martin Visbeck DEES, Lamont-Doherty Earth Observatory [email protected]

Energy to drive the Climate Earth’s Energy Budget, Solar Radiation, The Greenhouse Effect Martin Visbeck DEES, Lamont-Doherty Earth Observatory [email protected]

Nao Euclidiana

nao pergunte

Racismo nao

Nao julgueis-nao-julgados

Nao Conformidade

Entre 0 Estodo e 0 mercodo: 0 publico o nao-estotol*€¦ · Entre 0 Estado e 0 mercado: 0 publico nao-estatal • 17 ao nao-estatal enquanto produyao, nao incluindo 0 nao-estatal

NAO Programming using .NET and Webots 01-Introduction to NAO

Wave Climate and the North Atlantic Oscillationxtide.ldeo.columbia.edu/~visbeck/nao/poster/Woolf.pdf · gradient, and in time the close link between wave climate and the North Atlantic

Prof. Dr. Martin Visbeck Academic Education and Employment€¦ · Prof. Dr. Martin Visbeck Academic Education and Employment 1982 - 1989 Diploma Physical Oceanography, CAU/IfM Kiel

Miolo Nao gosto nao quero

Project Nao

General Circulation and Climate Zones Martin Visbeck DEES, Lamont-Doherty Earth Observatory [email protected]

NAO a new way to learn Meet NAO - State of Michigan€¦ · Meet NAO NAO a new way to teach writing and language arts: Students can write their own stories for NAO to tell or program

NAO Catalog

PROJECT NAO - Rutgers DCS Hacker Spacehackerspace.cs.rutgers.edu/library/NAO/NAO-programming.pdf · PROJECT NAO Conception ... Thanks to this project, ... Special thought for Mr J

Nao Victoria

Reporte Nao

Pra Nao Dizer Que Nao Falei Das Flores.musx

Biologo Nao

Ocean Stratification and Circulation Martin Visbeck DEES, Lamont-Doherty Earth Observatory [email protected]

Peter J. Lamb 1, 3, Mostafa El Hamly 1, 2, Rachid Sebbari ...xtide.ldeo.columbia.edu/~visbeck/nao/poster/Lamb.pdfPeter J. Lamb 1, 3, Mostafa El Hamly 1, 2, Rachid Sebbari 1, 2, Diane

Nao-Nao · 2021. 3. 1. · Nao-Nao Balancín. Garden swing Materiales. Materials Aluminio anodizado o termolacado. Lona de tela plástica. Gomaespuma de poliuretano. Tapicería desenfundable

YOTG Hamburg - Christiane Brandes-Visbeck - Talk to the Goat: Generation Future, Bro!

NAO ENTENDO